Vol. 114 No. 1 July 2012

Delayed-onset infections after impacted lower third molar extraction: involved and sensitivity profiles to commonly used antibiotics Rui Figueiredo, DDS,a Eduard Valmaseda-Castellón, DDS, PhD,b M. Florencia Formoso-Senande, DDS,c Leonardo Berini-Aytés, DDS, MD, PhD,d and Cosme Gay-Escoda, DDS, MD, PhD,e Barcelona, Spain University of Barcelona and Teknon Medical Center

Objectives. The objectives of this study were to identify the bacteria involved in delayed-onset infections after lower third molar removal and to determine the most suitable antibiotic for such complication. Study Design. Bacterial samples were collected from 13 patients who developed delayed-onset infections after lower third molar extraction. After the identification of the bacterial isolates, the in vitro antimicrobial susceptibility of the isolated strains was determined. Results. A total of 11 patients (12 samples) were finally included in the study. Up to 7 bacteria genera were identified. sp. was present in 11 patients, Prevotella sp. in 8 cases, and Peptostreptococcus sp. in 7. Some strains of these bacteria were not susceptible to amoxicillin, amoxicillin/clavulanate, and metronidazol, whereas no resistances were found to . Conclusions. Fusobacterium sp., Prevotella sp., and Peptostreptococcus sp. are frequently present in delayed-onset infections after lower third molar removal. Based on the results of the microbial susceptibility tests, clindamycin seems to be the most adequate antibiotic for the treatment of this complication. (Oral Surg Oral Med Oral Pathol Oral Radiol 2012;114:43-48)

The treatment of postoperative wound infections after Delayed-onset infections are a rare postoperative compli- lower third molar extraction has been widely discussed in cation with onset after suture removal (at least 7 days after dental literature. Most articles focus on the prevention of the surgical procedure), which can be difficult to man- such complications using systemic or local antibiotics.1-4 age.7,8 In fact, one previous report stated that antibiotics Usually, authors recommend amoxicillin associated with are effective in only two-thirds of the patients who devel- clavulanate as the first line of treatment.5,6 Nevertheless, oped this particular postoperative complication.8 Indeed, there are very few reports that study the involved bacteria, to treat this infection, surgical debridement of the extrac- as well as their sensitivity to commonly used antibiotics. tion site is often necessary, although some studies show that this rate is clearly inferior for conventional postoper- ative wound infections after lower third molar removal.8,9 This study was supported by a grant from the School of Dentistry of the University of Barcelona for PhD students. Therefore, the lack of data on the microbiological features aAssociate Professor of Oral Surgery, Professor of the Master’s of lower third molar postoperative infections, as well as degree program in Oral Surgery and Implantology, School of Den- the need to explain this low success rate of antibiotics, tistry of the University of Barcelona, Spain, Member of the IDIBELL justifies the need to perform a study with the following Research Group. aims: to identify the bacteria involved in delayed-onset bProfessor of Oral Surgery, Professor of the Master’s degree program in Oral Surgery and Implantology, School of Dentistry of the Uni- infections after lower third molar extractions and to de- versity of Barcelona, Spain, Member of the IDIBELL Research termine which is the most suitable antibiotic to treat such Group. complication. cMaster’s degree program in Oral Surgery and Implantology, School of Dentistry of the University of Barcelona, Spain. dProfessor of Oral and Maxillofacial Surgery, Professor of the Mas- MATERIAL AND METHODS ter’s degree program in Oral Surgery and Implantology, School of Bacterial samples of 13 consecutive patients who devel- Dentistry of the University of Barcelona, Spain, Member of the oped delayed-onset infections were collected. These pa- IDIBELL Research Group. tients had been submitted to an extraction of an impacted eChairman and Professor of Oral and Maxillofacial Surgery, Director of the Master’s degree program in Oral Surgery and Implantology, lower third molar in the Oral Surgery and Implantology School of Dentistry of the University of Barcelona; Oral and Max- Department of the School of Dentistry of the University of illofacial Surgeon, Teknon Medical Center, Barcelona, Spain, Coor- Barcelona, Spain. The main inclusion criterion was an dinator-researcher of the IDIBELL Research Group. inflammatory swelling of the operated area accompanied Received for publication Mar 28, 2011; returned for revision May 7, by the presence of suppuration that began at any time 2011; accepted for publication Jun 25, 2011. © 2012 Elsevier Inc. All rights reserved. subsequent to suture removal, 1 week postoperatively. 2212-4403/$ - see front matter The patients also had to be considered healthy (classifi- http://dx.doi.org/10.1016/j.tripleo.2011.06.022 cation ASA I or II). The exclusion criteria were the

43 ORAL AND MAXILLOFACIAL SURGERY OOOO 44 Figueiredo et al. July 2012 following: patients who took any antibiotic or antiseptic removal, tooth sectioning, the time elapsed from re- drug before collection of the bacterial sample; patients moval of the lower third molar to onset of the infection, who developed a surgical wound infection before suture the antibiotics prescribed to treat the infection, and the removal; and patients who presented periodontitis or deep need for an additional surgical procedure. Time and caries associated with the first and/or second adjacent date of the sample collection, the number of aerobic molars that could lead to a misdiagnosis. and anaerobic colony-forming units (CFU), the identi- fied bacteria, and the patient’s sensitivity to different Surgical technique antibiotics (amoxicillin, amoxicillin/clavulanate, clin- All patients had one lower third molar removed under damycin and ) were also recorded. local anesthesia—generally with a 4% articaine solution Two samples were excluded from the study be- containing epinephrine 1:100,000 (Artinibsa, Inibsa; Lliça cause of an inadequate shipping process (more than de Vall, Spain). The surgical technique used was similar 24 hours). to that described in a previous report.7 After the operation, the patients were prescribed an antibiotic (usually amoxi- Microbiology procedures cillin 750 mg every 8 hours for 7 days [Clamoxyl 750; To collect the bacterial samples, a sterile endodontic paper GlaxoSmithKline, Madrid, Spain], except in 1 patient point was inserted apically in the area where suppuration with previous history of allergy to penicillin who was was present (usually in the distal aspect of the adjacent prescribed clindamycin 300 mg every 6 hours for 7 days second molar) until resistance was encountered. It was left [Dalacin 300; Pfizer, Madrid, Spain]), a nonsteroidal anti- in place for 10 seconds and then inserted into a 2-mL inflammatory drug (usually sodium diclofenac 50 mg snap-top tube with reduced transport fluid medium every 8 hours [Diclofenaco Llorens 50 mg; Llorens; Bar- (RTF).13 Once collected, the samples were refrigerated celona, Spain]), an analgesic (metamizol 575 mg every 6 until shipped to the microbiology laboratory (Department hours for 3 to 4 days [Nolotil; Boehringer Ingelheim, Sant of Microbiology, Dentaid, Cerdanyola del Vallés, Spain) Cugat del Vallès, Spain]), and a mouthrinse (0.12% chlo- for inoculation and incubation. rhexidine digluconate every 12 hours for 15 days [Clo- After vortexing for 45 seconds, the samples were rhexidina Lacer; Lacer; Barcelona, Spain]). Postoperative 10-fold serially diluted in phosphate buffered saline instructions and use of the prescribed drugs were ex- (PBS; pH 7.2) and 100 ␮L of appropriated dilutions plained orally and also were provided on a printed sheet of were plated in duplicate on nonselective Columbia agar paper that was given to the patient. (Difco, Detroit, MI, USA) with 5% horse blood sup- plemented with vitamin K1 (10 mg/L) and hemin (5 Delayed-onset infection treatment mg/L). The samples were also plated on Dentaid-1 Bacterial samples were collected before performing any plates (Dentaid, Cerdanyola del Valles, Spain) for se- specific treatment for the delayed-onset infections. The lective isolation of Actinobacillus actinomycetemcomi- sockets were initially irrigated with chlorhexidine diglu- tans.14 Half of the blood agar plates were incubated in conate in a 0.12% solution, and the patients were pre- aerobic conditions for 3 days at 36°C and the other half scribed oral antibiotics (amoxicillin 875 mg plus clavu- in an anaerobic chamber for 14 days at 37°C with 80% lanate 125 mg every 8 hours) for 7 days. If this treatment N2, 10% CO2, and 10% H2. Dentaid-1 plates were did not resolve the infection, a second bacterial sample incubated in air plus 5% CO2 at 37°C for 5 days. was collected and the area was exposed by means of a Subsequently, the total count of aerobic and anaero- full-thickness flap. The granulation tissue and any bone bic bacteria was made. The most common pathogens particles or foreign material inside the extraction socket present in odontogenic infections were aisled and iden- were removed. The socket was then irrigated with sterile tified on the basis of their characteristic colony mor- saline, the flap was repositioned with 3-0 silk sutures phology; a Gram stain and an aerotolerance test were (Silkam; Braun, Tuttlingen, Germany), and clindamycin performed on each isolate. The anaerobic bacteria were 300 mg every 6 hours for 7 days was prescribed. identified using the Rap Id Ana II System (Remel, Oxford SA, Madrid, Spain) and Streptococcus were Data sampling identified using the Rapid STR System (Remel). Pure A single surgeon examined all patients. The following isolates were kept on plates, or were preserved at –70°C data were gathered: age, gender, operated side, smok- for posterior minimal inhibitory concentration (MIC) ing habit, history of pain or infection of the lower third determination. molar, position of the lower third molar according to the Winter classification,10 distal space and depth of Susceptibility testing inclusion using the Pell & Gregory classification,11 For the in vitro antimicrobial susceptibility testing, Nolla stage,12 degree of retention, flap design, bone antibiotic powders were used (amoxicillin, amoxicillin/ OOOO ORIGINAL ARTICLE Volume 114, Number 1 Figueiredo et al. 45

potassium clavulanate [provided by GlaxoSmithKline,

Madrid, Spain]; clindamycin [provided by MP Bio- Clavu ϩ medicals LLC, Illkirch, France], and metronidazole Ͻ 20 [provided by Sigma-Aldrich Quimica SA, Madrid, Spain]). Then, broth microdilution testing following the

guidelines described by the National Committee for Ͻ 20 Left Right No No TotalPartial Total Partial IIBHoriz. IIB Vert. Yes Yes Yes Yes No No Clinical Laboratory Standards15,16 was performed. The MICs of aerobic bacteria were determined by using Clavu Clind Amox 29 28 19 79 18 16 10 10 8 ϩ cation-adjusted Mueller-Hinton broth supplemented with #9 #10 #11 Ͻ 20 1% horse serum at 36°C for 24 hours. Reference strains of Left Yes Total Total IIIB Yes Yes Enterococcus faecalis ATCC29212 and Escherichia coli Yes ATCC 25922 were used as controls in each test. Clavu Amox 9

The MICs of anaerobic bacteria were determined by ϩ #8 16 71 using brain heart infusion broth (Difco) supplemented Right No Yes Total IIC Yes No with cistein (Merck, Darmstadt, Germany) 0.4 g/L, No ␮ ␮ horse serum 1%, vitamin K1 1 g/mL, hemin 5 g/mL in an atmosphere of 10% CO2, 10% H2, and 70% N2 at Clavu Amox ϩ #7 29 21 37°C for 72 hours. Reference strains of 10 Right No Total IIB Horiz. Horiz. Mes. Yes Yes fragilis ATCC 25285 and Bacteroides thetaiotaomi- No cron ATCC 29741 were used as controls in each test.

The interpretation of the MIC values was made using Clavu Amox 8 16 25 ϩ #6

the EUCAST (European Committee on Antimicrobial Ͻ 20 No Susceptibility Testing) breakpoints.17 Right Yes Total IIIB Vert. Yes Yes No

␤

-lactamase testing Clavu Amox 7 ϩ #5 16 Nitrocefin (Oxoid, SR0112) was rehydrated and the bac- 41 teria were tested emulsified into the nitrocefin drop onto a FFFFF MF Left No Total Partial Partial Partial Total IIIB Yes Yes clean glass slide. The result was considered positive if the No color changed from yellow to red. Bacteroides fragilis Clavu Amox 29 54 10 ϩ ATCC25285 was included as a positive control. #4 Ͻ 20 No M Right No IIC Yes Yes Statistical analysis No A descriptive analysis of the data was made with the Clavu Amox 31 45 10

Statistical Package for the Social Sciences (SPSS ver- ϩ #3 sion 15.0; SPSS, Chicago, IL). Ͼ 20 M Left Yes Partial Total Partial Total IIB Horiz. Horiz. Mes. Yes Yes No RESULTS Clavu Amox 6 17 A total of 11 patients (12 samples) were finally in- 28 ϩ #2 cluded in the analysis. The mean age of the patients was Ͼ 20 F Left No Total IIC Yes Yes 22.8 years and 7 were females. The mean time elapsed No from extraction to the delayed-onset infection was 38.7

days (range 16-79 days). Table I shows the main clin- Clavu Amox 9 ϩ #1 21 ical, radiological, and surgical features of the cases 28 M Left No No Total Partial Total Horiz. Mes. Yes Yes included in our sample. No Ten of the 11 patients were successfully treated with antibiotics. However, in case #9, the prescribed antimi- crobials were not effective and the infection was finally solved after surgical debridement (Table I). The isolated bacterial strains and the number of

CFUs are shown in Table II. The most common were Case

Fusobacterium sp. (present in 11 of the 12 samples); Clinical, radiographic, and surgical features of the patients Prevotella sp. (isolated in 8 samples), and Peptostrep- tococcus sp. (present in 7 samples). Table I. Amox, amoxicillin; Clavu, clavunalate; Clind, clindamycin; F, female; Horiz., horizontal; M, male; Mes., mesioangular; Vert., vertical. Age, y Gender Operated side Time from extraction – infection, d Smoking, cigarettes/d Previous infection Soft tissue coverage Bone coverage Pell & Gregory Classification IIB Winter Classification Nolla Bone removal Tooth sectioning Antibiotic prescribed after surgeryAntibiotic prescribed to treat infection Amox Amox Amox Amox Amox Amox Amox Amox Amox Amox Clind Amox The MIC values and sensitivity profiles of the bac- Surgical debridement ORAL AND MAXILLOFACIAL SURGERY OOOO 46 Figueiredo et al. July 2012

Table II. Aisled bacteria in each sample Case Age Gender CFU/mL aerobic CFU/mL anaerobic Isolated strains #1 21 Male 1.7 ϫ 105 5.5 ϫ 105 Veillonella sp. Prevotella intermedia #2 17 Female 2.1 ϫ 106 2.4 ϫ 107 Fusobacterium varium Peptostreptococcus micros Fusobacterium sp. Prevotella intermedia #3 31 Male 4.0 ϫ 106 1.3 ϫ 107 Peptostreptococcus prevotii Fusobacterium sp. Prevotella intermedia #4 29 Male 2.5 ϫ 106 1.3 ϫ 106 GPB aerobic Fusobacterium sp. Prevotella corporis #5 16 Female 2.6 ϫ 105 4.3 ϫ 105 Actinomyces israelii Fusobacterium nucleatum #6 16 Female 7.0 ϫ 105 2.3 ϫ 106 Fusobacterium sp. Prevotella intermedia #7 29 Female 3.88 ϫ 104 8.5 ϫ 104 Fusobacterium sp. Fusobacterium varium #8 16 Female 5.28 ϫ 106 1.9 ϫ 107 Fusobacterium sp. Prevotella intermedia Peptostreptococcus micros #9 29 Female 1.37 ϫ 104 1.8 ϫ 104 GPB aerobic Peptostreptococcus micros Fusobacterium sp. 7.7 ϫ 105 5.0 ϫ 106 Peptostreptococcus micros Fusobacterium sp. Prevotella intermedia Bacteroides forsythus #10 28 Male 1.2 ϫ 105 7.6 ϫ 105 Peptostreptococcus micros Prevotella intermedia #11 19 Female 7.2 ϫ 103 1.9 ϫ 104 Peptostreptococcus micros Porphyromonas endodontalis Fusobacterium sp. Note that patient # 9 required surgery in addition to antibiotics to treat the infection, so 2 samples were collected in this case. CFU, colony-forming units; GPB, gram-positive bacilli. teria are shown in Table III. Clindamycin showed ex- to be the cause of delayed-onset infections. Moreover, the cellent results for all isolated strains, followed by met- effect of postoperative antibiotics in this particular type of ronidazole and amoxicillin/clavulanate. Prevotella sp. late-onset complications is questionable. On the one hand and Fusobacterium sp. showed particularly high resis- because most patients had finished taking amoxicillin at tance rates to amoxicillin alone. least 3 weeks before diagnosis (the mean time elapsed from extraction to infection was 39 days), and on the other DISCUSSION hand, because the socket can be easily recontaminated To our knowledge, this is the first study to describe the with oral bacteria. microbiological characteristics of delayed-onset infec- It is commonly accepted that anaerobic bacteria tions after lower third molar extractions. This is a rare play a major role in the development of orofacial complication with an estimated incidence of 1.5%, infections.18 Nevertheless, there are very few reports which explains the reduced number of patients in our that actually try to identify the bacteria involved in series.7 Although the sample size may be considered surgical wound infections through microbiological small, the bacterial strains identified were quite similar sample collections. This information could be ex- in the great majority of patients. tremely useful to clinicians, as it allows a more Another limitation of our study is related to the fact that adequate prescription of antibiotics, especially in all patients were administered antibiotics after surgery. delayed-onset infections after lower third molar re- This could have selected the identified bacteria and also moval where antibiotics do not have a high success may have affected their sensitivity profiles. However, rate, as shown in a recent report.8 Our sample had these amoxicillin-susceptible microorganisms are unlikely very similar clinical and radiological features when OOOO ORIGINAL ARTICLE Volume 114, Number 1 Figueiredo et al. 47

Table III. MIC ranges and sensitivity profiles to amoxicillin, amoxicillin/clavulanic acid, clindamycin, and metro- nidazole of the most frequently identified bacteria genera Antibiotic Bacterial genera (n) MIC range, ␮g/mL S, % I, % R, % Amoxicillin Prevotella sp. (8) 2-32 0 12.5 87.5 Fusobacterium sp. (13) 0.0625 to Ͼ32 69.2 0 30.8 Peptostreptococcus sp. (7) Յ0.03125 to 0.25 100 0 0 Amoxicillin Clavulanate Prevotella sp. (8) 1-32 75 12.5 12.5 Fusobacterium sp. (13) Ͻ0.03125 to 32 69.2 7.7 23.1 Peptostreptococcus sp. (7) Ͻ0.03125 to 0.5/0.0045 to 0.072 100 0 0 Clindamycin Prevotella sp. (8) Ͻ0.0625 to 0.25 100 0 0 Fusobacterium sp. (13) Ͻ0.0625 to 0.25 100 0 0 Peptostreptococcus sp. (7) Ͻ0.0625 to 1 100 0 0 Metronidazole Prevotella sp. (8) Ͻ0.0625 to 0.25 100 0 0 Fusobacterium sp. (13) Ͻ0.0625 to Ͼ64 93.3 0 7.7 Peptostreptococcus sp. (7) 0.125 to Ͼ32 85.7 0 14.3 MIC, minimal inhibitory concentration; n, number of aisled strains; S, susceptible; I, intermediate; R, resistant. compared with previous articles,7,8,19 so it might be terial agents, our results showed that all strains of these expected that the microbiological profile is also com- 3 bacteria were susceptible to clindamycin. Kuriyama parable. et al.22 tested 800 anaerobic isolates found in dentoal- A mixed anaerobic-aerobic flora is frequently present in veolar infections and, like in our sample, found that odontogenic infections.5,20 Among these bacteria, viridans clindamycin had very low MIC to the great majority of group streptococci and staphylococci are usually pre- Fusobacterium, Peptostreptococcus, Porphyromonas, dominant. In our study, these bacteria might have and Prevotella strains. been eliminated owing to the administration of post- On the other hand, amoxicillin was shown to be an operative antibiotics and the use of chlorhexidine mouth- inadequate antibiotic for the treatment of these infections, rinses. Nevertheless, most authors attribute a causative as it presented extremely high MIC values to 2 of the role to other anaerobic bacteria, such as Prevotella, Bac- isolated strains. This could be attributable to the high teroides, Fusobacterium, or Peptostreptococcus.5,21,22 incidence of ␤-lactamase–producing bacteria from the ge- Our results entirely support this opinion, showing that nus Prevotella and Fusobacterium, as shown in several Fusobacterium sp. (present in 11 of the 12 samples ana- reports.22-24 A fact that supports this statement is that the lyzed), Prevotella sp. (found in 8 samples), and Pepto- sensitivity rates clearly improved with the addition of streptococcus sp. (present in 7 patients) were common in clavulanic acid. In fact, because of the sensitivity profiles our samples. In patient # 9, 2 samples were retrieved of the odontogenic infections, some authors recommend because the delayed-onset infection initially treated with the use of amoxicillin/clavulanate as the first line of treat- amoxicillin/clavulanate relapsed after 21 days. This case ment.5,25,26 In our sample, some strains of Prevotella sp. is of particular interest, as part of the microbiological flora and of Fusobacterium sp. were not susceptible to this were probably eliminated with the first antibiotic pre- association. Nevertheless, this combination can be a good scribed (amoxicillin/clavulanate). In the second sample, alternative to clindamycin. A previous clinical study Fusobacterium sp., Prevotella sp., and Peptostreptococcus claimed that amoxicillin/clavulanate and clindamycin had sp. were still present, which reinforces the hypothesis that similar results in the treatment of delayed-onset infections, these microorganisms could be the cause of this complica- being effective in two-thirds of patients.8 Metronidazole tion. The major risk factors for delayed-onset infections after can be considered a good option to treat postoperative lower third molar removal are total soft tissue coverage and infections especially when gram-negative anaerobic bac- a mesioangular or vertical angulation of the third molar.19 teria are involved. However, this antibacterial drug should Probably, these features allow a primary closure of the sur- be associated with another antibiotic, mainly because it gical wound, leaving a dead space beneath the mucosa, has a reduced effect over gram-positive aerobic bacteria.27 which makes oral hygiene measures (especially chlorhexi- The microbial sensitivity test results of this article dine mouthrinses) of that area ineffective. Furthermore, the seem to support the clinical data published in 2008, used flap design detaches the periodontal insertions of the where amoxicillin/clavulanate showed disappointing suc- adjacent second molar, which allows the penetration of bac- cess rates, with 33% of patients needing surgical debridement teria through the distal gingival sulcus of this tooth. All these of the extraction site.8 The microbiology outcomes of the aspects create an ideal environment for the development of present study seem to support the use of clindamycin in the anaerobic bacteria. treatment of delayed-onset infections after lower third molar Regarding the susceptibility to the different antibac- extraction. A future study with a larger sample and without ORAL AND MAXILLOFACIAL SURGERY OOOO 48 Figueiredo et al. July 2012 the use of postoperative antibiotics would be of great value to 14. Alsina M, Olle E, Frias J. Improved, low-cost selective culture confirm the present results. medium for Actinobacillus actinomycetemcomitans. J Clin Mi- crobiol 2001;39:509-13. 15. National Committee for Clinical Laboratory Standards. Methods CONCLUSIONS for dilution antimicrobial susceptibility tests for bacteria that Fusobacterium sp., Prevotella sp., and Peptostrepto- grow aerobically. Approved standard M7-A6. Wayne, PA; coccus sp. are frequently present in delayed-onset in- NCCLS; 2003. fections after lower third molar removal when postop- 16. National Committee for Clinical Laboratory Standards. Methods erative amoxicillin has been administered. Based on the for antimicrobial susceptibility testing bacteria that grow aero- bically. Approved standard M11-A6. 6th ed. Wayne, PA; results of the microbial susceptibility tests, clindamycin NCCLS; 2004. seems to be the most adequate antibiotic for the treat- 17. European Committee on Antimicrobial susceptibility testing; ment of this complication. Breakpoint tables for interpretation of MICs and zone diameters The authors thank Dentaid and in particular Dr. Ester Ollé v. 1.1.2010-04-27. Available at: http://www.eucast.org/clinical_ for help with the microbiological analysis of the samples. breakpoints. Accessed October 29, 2011. 18. Kuriyama T, Karasawa T, Nakagawa K, Saiki Y, Yamamoto E, Nakamura S. Bacteriologic features and antimicrobial suscepti- REFERENCES bility in isolates from orofacial odontogenic infections. Oral Surg 1. Bulut E, Bulut S, Etikan I, Koseoglu O. The value of routine Oral Med Oral Pathol Oral Radiol Endod 2000;90:600-8. antibiotic prophylaxis in mandibular third molar surgery: acute- 19. Figueiredo R, Valmaseda-Castellón E, Berini-Aytés L, Gay-Escoda C. phase protein levels as indicators of infection. J Oral Sci Delayed-onset infections after lower third molar extraction: a case- 2001;43:117-22. control study. J Oral Maxillofac Surg 2007;65:97-102. 2. Piecuch JF, Arzadon J, Lieblich SE. Prophylactic antibiotics for 20. Rajasuo A, Perkki K, Nyfors S, Jousimies-Somer H, Meurman third molar surgery: a supportive opinion. J Oral Maxillofac Surg JH. Bacteremia following surgical dental extraction with an 1995;53:53-60. emphasis on anaerobic strains. J Dent Res 2004;83:170-4. 3. Poeschl PW, Eckel D, Poeschl E. Postoperative prophylactic 21. Kuriyama T, Karasawa T, Nakagawa K, Yamamoto E, Naka- antibiotic treatment in third molar surgery—a necessity? J Oral mura S. Incidence of beta-lactamase production and antimicro- Maxillofac Surg 2004;62:3-8; Discussion:9. bial susceptibility of anaerobic gram-negative rods isolated from 4. Ren YF, Malmstrom HS. Effectiveness of antibiotic prophylaxis pus specimens of orofacial odontogenic infections. Oral Micro- in third molar surgery: a meta-analysis of randomized controlled biol Immunol 2001;16:10-5. clinical trials. J Oral Maxillofac Surg 2007;65:1909-21. 22. Kuriyama T, Williams DW, Yanagisawa M, Iwahara K, Shimizu 5. Poeschl PW, Spusta L, Russmueller G, Seemann R, Hirschl A, C, Nakagawa K, et al. Antimicrobial susceptibility of 800 anaer- Poeschl E, et al. Antibiotic susceptibility and resistance of the obic isolates from patients with dentoalveolar infection to 13 oral odontogenic microbiological spectrum and its clinical impact on antibiotics. Oral Microbiol Immunol 2007;22:285-8. severe deep space head and neck infections. Oral Surg Oral Med 23. Handal T, Olsen I, Walker CB, Caugant DA. Beta-lactamase Oral Pathol Oral Radiol Endod 2010;110:151-6. production and antimicrobial susceptibility of subgingival bac- 6. Lacasa JM, Jiménez JA, Ferrás V, Bossom M, Sóla-Morales O, teria from refractory periodontitis. Oral Microbiol Immunol García-Rey C, et al. Prophylaxis versus pre-emptive treatment 2004;19:303-8. for infective and inflammatory complications of surgical third 24. Mosca A, Miragliotta L, Iodice MA, Abbinante A, Miragliotta G. molar removal: a randomized, double-blind, placebo-controlled, Antimicrobial profiles of Prevotella spp. and Fusobacterium clinical trial with sustained release amoxicillin/clavulanic acid nucleatum isolated from periodontal infections in a selected area (1000/62.5 mg). Int J Oral Maxillofac Surg 2007;36:321-7. 7. Figueiredo R, Valmaseda-Castellón E, Berini-Aytés L, Gay- of southern Italy. Int J Antimicrob Agents 2007;30:521-4. Escoda C. Incidence and clinical features of delayed-onset in- 25. Brescó-Salinas M, Costa-Riu N, Berini-Aytés L, Gay-Escoda C. fections after extraction of lower third molars. Oral Surg Oral Antibiotic susceptibility of the bacteria causing odontogenic in- Med Oral Pathol Oral Radiol Endod 2005;99:265-9. fections. Med Oral Patol Oral Cir Bucal 2006;11:e70-5. 8. Figueiredo R, Valmaseda-Castellón E, Laskin DM, Berini-Aytés L, 26. Tomás I, Tomás M, Alvarez M, Velasco D, Potel C, Limeres J, Gay-Escoda C. Treatment of delayed-onset infections after impacted et al. Susceptibility of oral obligate anaerobes to telithromycin, lower third molar extraction. J Oral Maxillofac Surg 2008;66:943-7. moxifloxacin and a number of commonly used antibacterials. 9. White RP Jr, Shugars DA, Shafer DM, Laskin DM, Buckley MJ, Oral Microbiol Immunol 2007;22:298-303. Phillips C. Recovery after third molar surgery: clinical and 27. Maestre-Vera JR. Treatment options in odontogenic infection. health-related quality of life outcomes. J Oral Maxillofac Surg Med Oral Patol Oral Cir Bucal 2004;9;Suppl:S19-S31. 2003;61:535-44. 10. Winter GB. Impacted mandibular third molar. St. Louis: American Reprint requests: Medical Book; 1926. 11. Pell GJ, Gregory BT. Impacted mandibular third molars: classi- Rui Figueiredo, DDS fication and modified techniques for removal. Dent Dig 1933; Facultat d’Odontologia 39:330-8. Campus de Bellvitge (UB) Pavelló de Govern 12. Nolla C. Development of the permanent teeth. J Dent Child 2a planta, Despatx 2.9 1960;27:254-60. C/ Feixa Llarga s/n 13. Syed SA, Loesche WJ. Survival of human flora in E-08907–L’Hospitalet de Llobregat, Spain various transport media. Appl Microbiol 1972;24:638-44. [email protected]